Design and synthesis of 4-substituted indolo[3,2-e][1,2,3]triazolo[1,5-a] pyrimidine derivatives with antitumor activity

Article abstract of DOI:10.1021/jm700964u

New derivatives of the indolo[3,2-e][1,2,3]triazolo[1,5-a]pyrimidine system, substituted in the 4 position, were designed as novel antitumor agents because of their theoretical capability to form stable complexes with DNA fragments. The calculated free energies of binding were found in the range -12.76 → -39.68 Kcal/mol. The docking studies revealed a common binding mode with the chromophore intercalated between GC base pairs, whereas the side chain lies along the minor groove. Compounds, selected on the basis of the docking studies and suitably synthesized, showed antiproliferative activity against each type of tumor cell line investigated, generally in the low micromolar range. The more active derivatives were shown to be 1eJ and 1eL, endowed with significant antiproliferative activity against the renal and CNS subpanels, respectively. A mechanism of action closely related to the DNA-interacting drugs can be supposed, although, alternative mechanisms, similar to those of the anthracyclines, can also operate.

Full text of DOI:10.1021/jm700964u

J. Med. Chem. 2008, 51, 2037–2046
2037
Design and Synthesis of 4-Substituted Indolo[3,2-e][1,2,3]triazolo[1,5-a]pyrimidine Derivatives
with Antitumor Activity
Antonino Lauria, Chiara Patella, Gaetano Dattolo, and Anna Maria Almerico*
Dipartimento Farmacochimico, Tossicologico e Biologico - UniVersità di Palermo, Via Archirafi 32, 90123 Palermo, Italy
ReceiVed August 2, 2007
New derivatives of the indolo[3,2-e][1,2,3]triazolo[1,5-a]pyrimidine system, substituted in the 4 position,
were designed as novel antitumor agents because of their theoretical capability to form stable complexes
with DNA fragments. The calculated free energies of binding were found in the range –12.76 f -39.68
Kcal/mol. The docking studies revealed a common binding mode with the chromophore intercalated between
GC base pairs, whereas the side chain lies along the minor groove. Compounds, selected on the basis of the
docking studies and suitably synthesized, showed antiproliferative activity against each type of tumor cell
line investigated, generally in the low micromolar range. The more active derivatives were shown to be 1eJ
and 1eL, endowed with significant antiproliferative activity against the renal and CNS subpanels, respectively.
A mechanism of action closely related to the DNA-interacting drugs can be supposed, although, alternative
mechanisms, similar to those of the anthracyclines, can also operate.
Introduction
derivatives. The intermediate resulting from the cycloaddition
reaction is a 1-(2-carboxyethyl-heteroaryl)-2-aminotriazole, which
presents an amino group able to give rise to the pyrimidine ring,
being within reacting distance to the carboxylate function. We have
already shown that this type of reaction can be used in the pyrrole/
indole series and have demonstrated that the nature of the substrate
and the reaction conditions can widely influence the nature of the
reaction products. Although, in the presence of the same type of
complexity (two rings formed in sequence), for this type of domino
reaction, the bond-forming economy greatly depends on the nature
of the starting material (pyrrole/indole) and on the substitution
pattern. However, a minor modification of the experimental
procedure allows the reaction to become suitable for a general
application in both series. By this approach we were able to
synthesize derivatives of several new systems such as (pyrrolo[3,4-
e] and pyrrolo[2,3-e])[1,2,3]triazolo[1,5-a]pyrimidine and (in-
dolo[2,3-e] and indolo[3,2-e])[1,2,3]triazolo[1,5-a]pyrimidine.7–10
DNA represents one of the more important molecular cellular
targets of several chemotherapeutic drugs. Molecular recognition
of DNA by small molecules and proteins is a fundamental
problem in drug design. Polycyclic heterocycles having a planar
structure can be effective pharmacophore moieties of DNA-
interactive drugs because they can insert between the stacked
base paired oligonucleotides. Moreover, if they bear suitable
side chains, further interactions of these ligands with the other
important architectural feature of DNA, its minor groove, can
be envisaged.
Among the different classes of antitumor drugs that interact
with DNA, the actinomycins, a family of chromopeptide
antibiotics, represent a peculiar one that combines the two
above-mentioned features. The most representative derivative,
Dactinomycin (Actinomycin D), has been extensively investi-
gated and its sequence specificity has been analyzed in detail
by a variety of methods, including chemical footprinting,1
NMR,2 X-ray crystallography,3 and photoaffinity cross-linking.4,5
It binds to double-stranded DNA and the phenoxazone moiety
intercalates between bases, whereas the peptide substituents lie
in the minor groove.6 The biological activity appears to depend
on the very slow rate of dissociation of the complex between
DNA and the drug, which reflects the intermolecular hydrogen
bonds, the planar interaction between the purine rings and the
chromophore, and the numerous van der Waals interactions
between the polypeptide side chain and the DNA.
Some of the derivatives belonging to these classes of
polycycles were selected for screening tests by the National
Cancer Institute (NCIa) in the Developmental Therapeutic
Program (DTP). In the primary anticancer screening assay, tested
against a three-cell line panel consisting of MCF7 (breast), NCI-
H460 (lung), and SF-268 (CNS), most of the annelated triazolo-
pyrimidines were inactive up to 10-4 M concentration. However,
considering that, in the case of the antitumor class of actino-
mycins, the peptide chains play a fundamental role in the DNA
interacting capability, we decided to design new series of
potential DNA-interactive compounds using, as template for the
pharmacofore planar moiety, one of the condensed triazolo-
pyrimidine systems in which several side chains can be
incorporated. Therefore, we report herein the design, synthesis,
In this context, we have already been interested in the design
and development of several types of tricyclic/tetracyclic derivatives
presenting a triazolo-pyrimidine moiety further condensed with a
pyrrole/indole ring,7–10 stucturally correlated to classes of DNA-
intercalating agents. The synthetic approach to these new hetero-
cyclic ring systems was exhaustively investigated by us during these
studies: it always involved a domino reaction between an azido-
pyrrole/indole-ortho-carboxylate and substituted acetonitriles. The
azido moiety acts as a 1,3-dipolar reagent in cycloaddition reactions
with dipolarophiles such as anions obtained from methylene active
a Abbreviations: NCI, National Cancer Institute; DTP, Developmental
Therapeutic Program; CNS, central nervous system; PDB, Protein Data
Bank; LGA, Lamarckian genetic algorithm; DMF, N,N-dimethylformamide;
TEA, triethylamine; DCC, dicyclohexylcarbodiimide; DCI, 1,1′-carbonyl-
diimidazole; DCM, dichloromethane; MG_MID, mean graph midpoint;
PCC, Pearson correlation coefficient; 3D-MIND, Drug Discovery and Data
Mining Information for New Directions; SOMs, self-organizing maps.
* To whom correspondence should be addressed. Tel.: +39-0916161606.
Fax: +39-0916169999. E-mail: almerico@unipa.it.
10.1021/jm700964u CCC: $40.75
2008 American Chemical Society
Published on Web 03/18/2008

2038 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 7
Table 1. Selected Side Chain Structures
Lauria et al.
and biological screening results on new 4-substituted indolo[3,2-
e][1,2,3]triazolo[1,5-a]pyrimidine derivatives.
the best ligand–receptor binding arrangement. Docking to DNA
fragment was performed using the Lamarckian genetic algorithm
(LGA)14 and an empirical binding free energy function, previ-
ously tested with success by us in the reconstruction of known
DNA-ligand complexes.15 The PDB was searched for DNA
fragments bound with intercalators and the structure 1DSC (an
octamer complexed with Actinomycin D) was selected. The
original ligand was removed and the DNA sequence was utilized
for the docking experiments.
In the case of derivatives 1a-e, the change of free energy of
binding was found in the range -11.28 f –13.79 Kcal/mol
(cfr Table 2, row A) and is more negative than the value
obtained by us, in these calculations, for Actinomycin D (-10.37
kcal/mol). Therefore, these ∆G values calculated for this class
of tetracyclic heterocycles demonstrated that, although not being
active in the in vitro tests carried out so far, derivatives of this
system possessed the electronic and steric feature to interact
with DNA and could constitute suitable lead compounds for
new anticancer agents. In fact, if they bear suitable side chains
in the 4 position of the ring, further interactions of these ligands
with DNA minor groove can be envisaged.
Results and Discussion
Molecular Modeling. First, selected derivatives of the ring
system indolo[3,2-e][1,2,3]triazolo[1,5-a]pyrimidine, compounds
1a-e, were tested by docking experiments. The starting point
of all docking calculations is generally the crystal structure of
a macromolecule (enzyme, DNA, or RNA fragments and so
on), usually obtained from a macromolecule-ligand complex.
This complex may be taken from a database of compounds,
such as the Cambridge Crystallographic Database11 or the
Protein Data Bank (PDB).12 The primary consideration in the
design of a docking process has to be the development of
methods that are both rapid and reasonably accurate.
At present, relatively little is understood about the mode of
action at the molecular level of the majority of minor groove-
interacting drugs, although, there is an increasing evidence that
they may act by directly blocking or inhibiting protein-DNA
recognition. A total of 21 chain shape moieties B-V of variable
length and with different physicochemical character were
selected (Table 1). All the related compounds, obtained by
the insertion of these chains on the 4 position of derivatives
1a-e were used in the docking calculation to study the types
of interaction with DNA.
According to the procedure reported in the Experimental
Section, our studies were carried out by using the software
package AutoDock.13 This one combines a rapid energy
evaluation through precalculated grids of affinity potentials with
a variety of search algorithms to find suitable binding positions
for a ligand on a given receptor. While the macromolecule is
required to be rigid, the program allows torsional flexibility in
the ligand. The program searches for the best conformation and
for the best place of binding of the ligand within the receptor
structure. Several runs of space search were carried out to find
The predicted binding free energy of the lowest energy
docked structure for each ligand considered in this study is
reported in Table 2. The calculated ∆G values are in the range

Indolo[3,2-e][1,2,3]triazolo[1,5-a]pyrimidine
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 7 2039
Table 2. Docking Resultsa
∆G value more negative than the average one (calculated for
the same chain).
1a
1b
1c
1d
1e
mean
Inallthecases,the4-substitutedindolo[3,2-e][1,2,3]triazolo[1,5-
a]pyrimidine derivatives were shown to assume a spatial
arrangement in which the planar heterocyclic moiety intercalates
in GC portion of DNA sequence, whereas the side chain lies
close to the minor groove. These findings are exemplified in
Figure 1 (right), which shows the DNA-binding mode for the
most interesting derivative, presenting the chain L, the more
efficient in enhancing the DNA binding capability, linked to
the phenyl-substituted lead compound 1e, that has shown the
best DNA affinity (∆Gbinding ) -39.68 Kcal/mol).
Chemistry. With these premises, the synthesis of the
compounds, selected on the basis of the molecular modeling
results, [1e(I, J, K, L, N, Q, R)] was successfully achieved
through suitable steps, as shown in Scheme 1. The starting
tetracycle 1e was prepared through the classical approach
involving the domino reaction between the ethyl 2-azido-1-
methyl-1H-indole-3-carboxylate and benzylcyanide.9
A
B
C
D
E
F
G
H
I
-11.28 -12.09 -13.79 -12.66 -13.76 -12.72
-12.90 -14.12 -15.51 -14.46 -15.49 -14.50
-13.83 -14.39 -15.99 -15.23 -16.40 -15.17
-16.32 -16.23 -17.49 -16.58 -18.36 -17.00
-12.76 -13.98 -15.59 -15.09 -15.98 -14.68
-14.52 -14.35 -15.93 -15.52 -16.07 -15.28
-16.42 -15.95 -18.28 -17.21 -17.59 -17.09
-12.96 -14.14 -15.64 -15.29 -16.08 -14.82
-16.39 -17.62 -18.68 -18.08 -19.04 -17.96
-19.52 -20.70 -20.96 -20.44 -20.49 -20.42
-18.30 -17.89 -19.95 -18.58 -20.07 -18.96
-33.44 -33.91 -36.76 -34.79 -39.68 -35.72
-16.47 -16.86 -18.34 -17.31 -18.35 -17.47
-17.65 -17.99 -19.57 -16.79 -19.77 -18.35
-14.76 -14.92 -16.13 -15.74 -15.96 -15.50
-14.65 -14.17 -16.60 -15.83 -16.83 -15.62
-24.99 -27.85 -27.41 -26.95 -26.71 -26.78
-26.73 -30.21 -30.96 -28.61 -33.86 -30.07
-14.37 -14.93 -16.22 -15.94 -16.73 -15.64
-14.51 -15.68 -16.33 -15.49 -15.45 -15.49
-17.09 -16.02 -17.00 -18.50 -19.46 -17.61
-17.43 -17.15 -18.70 -18.14 -18.07 -17.89
J
K
L
M
N
O
P
Q
R
S
T
U
V
Derivative 1eK was synthesized in two steps: by reaction of
1e and 1-bromo-3-chloropropane, the chloro derivative 2 was
obtained. This last derivative, upon heating under reflux in
dibutylamine solVent free, gave 1eK in good yield.
MEAN -17.15 -17.78 -19.22 -18.37 -19.60
a All energies (∆Gbinding) are given in Kcal/mol.
-12.76 f -39.68 Kcal/mol and it is possible to observe a
drastic variation of ∆Gbinding if compared with the values
obtained for the lead compounds A. Analyzing the results for
the different chains, a regular increase of the binding affinity
was generally observed in homologous series for each type of
derivative (a-e) and in the average, compare for example chain
E and H or chain U and V, for which a mean gain of 0.14–0.18
Kcal/mol was generally evidenced, whereas the effect was more
evident in the case of the ester derivatives (0.88 Kcal/mol, chain
M vs N). The same effect was more enhanced if the chain
elongation is carried out on the amino end; thus, a net gain of
0.6–0.7 Kcal/mol was observed when switching from the free
amino end (chains B or E) to the methylamino (chains C or F)
up to the dibutylamino compounds (chains D or G). And,
eventually, for these series, the best ligands were found to be
those in which both effects were combined, giving rise to
compounds presenting the chain K, with a net increase in the
∆Gbinding values of about 4 Kcal/mol. The influence of the
peptido moiety, presenting different acid/basic ends in the side
chains, was also investigated (chains J, Q, R, and L). It was
possible to evidence a general increase in the binding affinity
with respect to the derivatives presenting the chain K. A
histidine end is usually more effective in enhancing the groove
interaction (compare for example chain J and Q), with a net
increase of ≈6.4 Kcal/mol in ∆Gbinding values. Such an effect
became more evident if a bis-peptide (glycine moiety) is
introduced to lengthen the chain (compare Q and R derivatives),
for which a ∆Gmean ) -30.07 Kcal/mol was calculated.
However, on the whole, the best chain resulted to be L which,
combining all the more interesting features above evidenced,
conferred to the polycycles of type 1 the highest DNA-binding
capability in each case and on the average (∆Gbinding in the range
-33.44 f -39.68 Kcal/mol). Thus, as an overall result, among
the side chains, I, J, K, L, N, Q, and R types conferred to the
compounds in which they were incorporated the more negative
energy of binding expressed as ∆Gmean values. With respect to
the nature of the lead compounds, on the average, the one with
the highest affinity was demonstrated to be derivative 1e (∆Gmean
) -19.60 Kcal/mol); in this case, the insertion of the above-
reported chains generally led to derivatives with a calculated
The majority of compounds bearing the suitable chains were
prepared through the key intermediate 1eN. Thus, this derivative
was first obtained from the sodium salt of 1e, generated in situ
by using sodium hydride, and ethyl 4-bromobutyrate in dry
DMF, but in very low yield. The use of potassium carbonate
(5-fold excess of the starting material), in dry acetonitrile,
increased the yields of 1eN to 70%. Subsequent hydrolysis with
NaOH in EtOH/H2O mixture quantitatively yielded the corre-
sponding carboxylic acid 1eI.
Derivative 1eL was synthesized by reacting 1eI with glycine
ethyl ester, in dry dichloromethane and triethylamine (TEA),
and using dicyclohexylcarbodiimide (DCC) as coupling agent.
Hydrolysis of 1eL led to the carboxylic acid 3, which was
further reacted with L-histidine methyl ester dihydrochloride,
under the above-reported coupling conditions, to give derivative
1eR in 50% yield.
The amide derivative 1eQ was obtained from the same key
acid 1eI upon reaction with L-histidine methyl ester dihydro-
chloride, in dry dichloromethane and TEA, and using DCC as
coupling agent.
In all the reactions carried out with DCC, compounds 4 and
5 were isolated and characterized (Scheme 2). To these side
products the structure of N-acylureas was assigned, ruling out
the isomeric O-acylisoureas 4′ and 5′, because the treatment of
isolated specimen with histidine did not result in any conversion
into 1eR or 1eQ. These findings resulted in an agreement with
the behavior observed by several researchers16 and with the
reactivity of carbodiimides.17 However, the two N-acylureas
were considered as possible DNA binders and, therefore, they
were evaluated by docking calculations, showing a binding
capability comparable with all the previously selected com-
pounds (∆Gbinding ) -29.05 Kcal/mol for 4 and -27.07 Kcal/
mol for 5).
Therefore, to avoid the formation of 4 or 5, the reactions of
3 or 1eI with L-histidine methyl ester dihydrochloride were
carried out in the presence of a different coupling agent (1,1′-
carbonyldiimidazole, DCI). By using these experimental condi-
tions, the 4-substituted derivatives were isolated in 70–80%
yields.

2040 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 7
Lauria et al.
Figure 1. Stick view of 10 superpositioned best scores (left) and best structure docked to DNA fragment (right) for derivative 1eL.
Scheme 1a
a Reagents and conditions: (i) K2CO3/MeCN, Cl(CH2)3Br; (ii) NHBu2; (iii) K2CO3/MeCN, ethyl 4-bromobutyrate; (iv) NaOH, EtOH/H2O; (v) (1) DCC/
DCM, 0 °C, (2) glycine ethyl ester hydrochloride/TEA; (vi) (1) DCC/DCM, 0 °C, (2) L-histidine methyl ester dihydrochloride/TEA; (vii) (1) DCI/dioxane,
(2) histamine.
Compound 1eJ was suitably prepared by amidation of the
acid 1eI with histamine, in dioxane, in the presence DCI as
coupling agent.
nine disease subpanels, including leukemia, nonsmall-cell lung,
colon, central nervous system, melanoma, ovarian, renal,
prostate, and breast tumors cell lines (Table 4).
Biological Activity. All the new 4-substituted indolo[3,2-
e][1,2,3]triazolo[1,5-a]pyrimidine derivatives 1e (I, J, K, L, N,
Q, and R) and the intermediates 4 and 5 were submitted to the
NCI for testing in the DTP.
Three derivatives (1eN, 1eI, and 1eQ) were tested according
to the prescreening procedure against three cell lines, MCF7
(breast), NCI-H460 (lung), and SF-268 (CNS), were found to
be inactive at 10-4 M concentration (Table 3) and, therefore,
were not screened further.
The in vitro test system and the information, encoded by the
activity pattern over all cell lines, were obtained (see Experi-
mental Section) according to the previously reported methodol-
ogy.18 The antitumor activity of a test compound is given by
three parameters for each cell line: pGI50 value (GI50 is the molar
concentration of the compound that inhibits 50% net cell
growth), pTGI value (TGI is the molar concentration of the
compound leading to total inhibition of net cell growth), and
pLC50 value (LC50 is the molar concentration of the compound
that induces 50% net cell death). Moreover, a mean graph
midpoint (MG_MID) is calculated for each of the mentioned
parameters, giving an average activity parameter over all cell
Instead, all the other derivatives 1e (J, K, L, and R) and the
N-acylureas 4 and 5 were directly tested against a panel of
approximately 60 tumor cell lines that have been grouped in

Indolo[3,2-e][1,2,3]triazolo[1,5-a]pyrimidine
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 7 2041
Scheme 2. Chemical Structures of O-Acylisoureas and
N-Acylureas
75 cell line (pGI50 ) 5.95), which was the type of tumor cell
line most susceptible to this class of heterocycles. Derivative
1eK furnished good response toward most of the cell line
belonging to the renal subpanel, and especially, in the case of
TK-10, for which a pGI50 ) 5.72 was calculated. In the case of
the N-acyl-dicyclohexylurea derivatives 4 and 5, it is possible
to evidence that 5 was only specifically active against the
majority of CNS and breast cancer lines, although, the higher
effect was found against IGROV1 of the ovarian panel (pGI50
) 4.81), whereas derivative 4 showed a wider range of cytotoxic
effects. In this case, it is also necessary to underline an
anomalous behavior against SNB-19 (CNS cancer) and SK-
OV-3 (ovarian cancer) for which a strong inhibitory effect was
evidenced at micromolar concentrations, although, it was not
possible to quantify the pGI50 value according to the standard
NCI procedure.
When valuating this class of compounds with respect to the
MG_MID values at TGI and LC50 level (Table 6), again
derivative 1eJ was demonstrated to be the most active as far as
the MG_MID values are concerned, although, the renal and CNS
subpanel cell lines resulted sensitive for compound 1eL, as
evidenced by the panel MG_MID values (4.68 and 4.82,
respectively) higher than the overall cell lines one. It is also of
remarkable interest the low number of cell lines giving positive
LC50 in the case of 1eL and 4, a result that can give account of
the low toxicity of such compounds.
Table 3. Growth Percentage of Tumor Cell Linesa
cmpd
MCF7
NCI-H460
SF-268
1eN
1eI
1eQ
114
80
116
95
95
117
110
87
128
a Data obtained from NCI’s in vitro disease-oriented tumor cell screen.
With respect to Dactinomycin (our starting reference drug),19
the series of 4-substituted indolo-triazolo-pyrimidines resulted
less potent, but our more interesting compounds, 1eJ and 1eL,
showed selectivity with respect to renal and CNS subpanel cell
lines against which Dactinomycin was not particularly effective.
lines. For the calculation of the MG_MID, insensitive cell lines
are included with the highest concentration tested. The discovery
of compounds with new selectivity patterns is one of the targets
of the screening program. Selectivity of a compound with respect
to a certain cell line of the screen is characterized by a high
deviation of the particular cell line parameter compared to the
MG_MID value.
An evaluation of the data reported in Table 4 revealed that,
with the exception of compound 1eR, active only against the
IGROV1 cell line (pGI50 ) 4.14) and PC-3 (pGI50 ) 4.52), all
the new synthesized indolo[3,2-e][1,2,3]triazolo[1,5-a]pyrim-
idines showed antiproliferative activity against the majority of
human tumor cell lines investigated, generally in the low
micromolar range. In fact, the number of sensitive cell lines
was 39/56 for 1eL, 49/52 for 4, 51/55 for 1eK, and 54/56 for
1eJ (Table 5), and only in the case of compound 5 the number
dropped to 15/59.
Considering the MG_MID values, in the range 4.06-5.04
(Table 5), the most active compound of the series was
demonstrated to be derivative 1eJ, at either GI50 and TGI50 level
(5.04, 4.67; see Supporting Information), and for the number
of responding cell lines, followed by 1eL, 4, and 1eK.
With respect to the tumor subpanel, the indolo[3,2-
e][1,2,3]triazolo[1,5-a]pyrimidines resulted particularly effective
against all kind of leukemia and, in particular, against the CCRF-
CEM cell line (pGI50 in the range 4.58-5.82). In fact, the
calculated pGI50 MG_MID values of leukemia subpanel are
always higher than the overall cell lines MG_MID values, with
the exception of 1eL. However, this last showed an excellent
response for CNS cancer subpanel. In the case of the most active
derivate 1eJ, it is also evident the good response of all renal
subpanel cell lines with an average panel MG_MID value )
5.17, higher than the overall cell lines one, although, the most
sensitive cell line resulted to be HOP-92 of the lung subpanel,
with pGI50 ) 5.87.
To predict the mechanism of action, the computerized analysis
COMPARE20 was performed on the most active derivative 1eJ
(MG_MID pGI50 ) 5.04). Tested as seed against the NCI
“Standard Agents” database, it showed a Pearson correlation
coefficient (PCC) of 0.551 at GI50 level and a lower value at
TGI level (0.542). In all the cases, the first rank was with
Acodazole (NSC305884), a typical intercalating agent.
We also utilized the facility 3D-MIND (Drug Discovery and
Data Mining Information for New Directions)21 tools that allows
a simultaneous examination of the information found in the DTP
antitumor drug screen by using “self-organizing maps” (SOMs),
which cluster this data in the high-dimensional GI50 space and
provide a means of its visual translation into a two-dimensional
anticancer map. SOM anticancer maps organize the data from
tested agents into regions, which share the same pattern of
growth inhibition and which substantially reflect their molecular
targets and modes of action. The Projected Cluster List gives
the NSC number and the location corresponding to the projected
compound. The results of this analysis on our derivative indicate
its location in the N2 region map, related to membrane integrity.
Moreover, by analyzing the mechanism of action map, 1eJ was
found located between the cytidine analogues and the anthra-
cyclines and, in particular, in the region in which the 1,4-bis[[2-
[(2-aminoethyl)amino]ethyl]amino]-5,8-dihydroxy-9,10-an-
thracenedione (NSC660224) is found.
These results, together with the low value of the PCC in the
COMPARE analysis, could suggest that this new class of
compounds may have a mechanism of action that could be
related to that of pure intercalating drugs, although, the behavior
as antimetabolite agents can not be excluded. We suppose that
both the planar chromophore and the side chain are fundamental
for the interaction with DNA and that this interaction may help
The inhibition shown by the derivative 1eL against the CNS
subpanel was remarkable, in particular, with respect to the SNB-

2042 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 7
Lauria et al.
Table 4. Inhibition of In Vitro Cancer Cell Lines by 4-Substituted Indolo-triazolo-pyrimidinesa
b
b
pGI50
pGI50
panel/cell line
1eJ
1eL
4
1eK
5
1eR
panel/cell line
SK-MEL-2
SK-MEL-28
SK-MEL-5
UACC-257
UACC-62
1eJ
1eL
4
1eK
5
1eR
Leukemia
4.58 5.08
4.69 5.54
5.33 4.64
<4.30 4.84
4.97 5.37
Melanoma
5.02 4.48
<4.30 nd
nd
CCRF-CEM
HL-60(TB)
K-562
MOLT-4
RPMI-8226
SR
5.82
4.93
5.05
5.40
4.98
5.39
5.38
5.59
4.86
4.86
5.22
nd
<4.00 <4.00
<4.00 <4.00
<4.00 <4.00
<4.00 <4.00
<4.00 <4.00
<4.00 <4.00
4.99
5.05
5.04
4.92
5.01
<4.00 <4.00 <4.00
<4.00 <4.00 <4.00
4.50
4.54
4.65
5.22
<4.00 <4.00
<4.00 <4.00
<4.00 <4.00
<4.30 4.18
5.14 4.59
Ovarian Cancer
5.05 5.55
4.90 4.71
<4.30 nd
4.56 4.81
4.60 5.13
5.17 nc
Renal Cancer
5.29 4.73
5.14 4.51
5.03 4.58
5.08 4.49
5.88 4.96
<4.30 4.71
5.06 <4.00 5.72
4.32 4.50
CNS Cancer
5.07 4.64
4.49 4.73
Breast Cancer
4.50 5.44
ndc
5.52
Non-Small Cell Lung Cancer
IGROV1
nd
nd
4.81
<4.00 <4.00
4.42 <4.00
<4.00 <4.00
<4.00 <4.00
<4.00 <4.00
4.14
A549/ATCC
EKVX
HOP-62
4.89
5.00
5.12
5.87
<4.30
4.88
4.96
4.96
nd
4.48 4.42
4.69 nd
5.23 4.82
5.44 nd
4.60 4.40
4.93 4.81
<4.30 <4.00 4.54
<4.30 4.49
5.19 4.80
Colon Cancer
<4.30 5.55
<4.30 4.99
5.04 4.85
<4.30 4.97
<4.30 nd
<4.30 4.50
<4.30 5.42
CNS Cancer
4.85 4.82
4.36
4.25
4.51
4.73
4.24
4.89
<4.00 <4.00
<4.00 <4.00
<4.00 <4.00
<4.00 <4.00
<4.00 <4.00
<4.00 <4.00
<4.00 <4.00
<4.00 <4.00
4.16 <4.00
OVCAR-3
OVCAR-4
OVCAR-5
OVCAR-8
SK-OV-3
5.08
4.99
4.96
5.00
4.99
4.31
4.76
4.65
4.40
4.52
HOP-92
NCI-H226
NCI-H23
NCI-H322M
NCI-H460
NCI-H522
786–0
A498
5.66
5.12
5.01
4.92
5.13
5.06
5.32
5.13
4.41
4.37
4.46
4.30
4.85
4.71
<4.00 <4.00
4.18 <4.00
<4.00 <4.00
<4.00 <4.00
4.20 <4.00
<4.00 <4.00
<4.00 <4.00
<4.00 <4.00
4.39
nd
ACHN
CAKI-1
RXF 393
SN12C
TK-10
UO-31
COLO 205
HCC-2998
HCT-116
HCT-15
HT-29
5.07
5.02
5.23
4.84
5.07
5.09
4.89
5.30
4.69
4.73
4.68
5.11
4.47
4.58
<4.00 <4.00
<4.00 <4.00
4.04 <4.00
<4.00 <4.00
<4.00 <4.00
<4.00 <4.00
<4.00 <4.00
5.22
KM12
SW-620
PC-3
DU-145
5.05
5.16
4.74
4.32
<4.00
<4.00 <4.00
4.52
SF-268
SF-295
SF-539
SNB-19
SNB-75
U251
4.93
5.04
5.03
4.93
4.98
5.02
<4.00
4.01 <4.00
4.10 <4.00
4.42 <4.00
MCF7
NCI/ADR-RES
MDA-MB-231/ATCC 5.03
HS 578T
MDA-MB-435
T-47D
4.89
4.43
4.85
4.44
4.40
4.41
4.26
nd
4.19 <4.00
4.02 <4.00
4.14 <4.00
<4.00 <4.00
<4.00 <4.00
4.25 <4.00
4.14 <4.00
5.32 <4.00 4.50
<4.30 <4.30 4.81
5.37 4.70
4.18
4.83 4.91
5.35 5.03
<4.30 4.60
4.36 nd
5.23 ncd
<4.00 <4.00 <4.00
4.82
4.48
4.98
4.99
4.98
nd
5.95 nd
4.35 <4.00
<4.00 <4.00
5.38 4.91
Melanoma
4.60 4.88
<4.30 4.33
4.73 4.51
BT-549
nd
4.57
LOX IMVI
MALME-3M 5.12
M14 4.92
4.95
4.55
4.55
4.67
<4.00 <4.00
<4.00 <4.00
<4.00 <4.00
MG_MIDe
5.04
4.79 4.78
4.61
4.06
4.01
a Data obtained from NCI’s in vitro disease-oriented tumor cells screen. b pGI50 is the –log of the molar concentration causing 50% growth inhibition of
tumor cells. c Not determined. d Not calculated. e MG_MID ) mean graph midpoint ) arithmetical mean value for all tested cancer cell lines. If the indicated
effect was not attainable within the used concentration interval, the highest tested concentration was used for the calculation.
Table 5. Overview of the Results (pGI50) of the In Vitro Antitumor
compounds worthy of further development. Because the COM-
PARE analysis and 3D-MIND tools revealed a mechanism of
action strictly related to that of DNA-interacting drugs, but not
ruled out alternative ones, the investigation and discovery of
the exact mechanism can constitute a decisive step toward the
optimization of the antitumor activity.
Screening for Indolo-triazolo-pyrimidinesa
b
# cell lines
investigated
# cell lines giving
positive pGI50
pGI50
range
cmpd
MG_MIDc
5
1eK
4
1eL
1eJ
59
55
52
56
56
15
51
49
39
54
4.81–4.01
5.72–4.18
5.55–4.18
5.95–4.32
5.87–4.84
4.06
4.61
4.78
4.79
5.04
Experimental Section
a Data obtained from the NCI’s in vitro disease-oriented human tumor
cells screen. b pGI50 is the -log of the molar concentration that inhibits
50% net cell growth. c MG_MID ) mean graph midpoint ) arithmetical
mean value for all tested cancer cell lines. If the indicated effect was not
attainable within the used concentration interval, the highest tested
concentration was used for the calculation.
Molecular Modeling. Ligand Setup. The 3D structures of the
ligands were constructed, preoptimized with implemented MM2
force field, and further refined using Vega 1.5.22 The PM3 method
was used for geometry optimization and charge calculation. To
allow flexibility in the ligand, it is necessary to assign the routable
bonds; in our case, all the rotamers were allowed.
DNA Fragment Setup. The PDB file of the selected structure
1DSC was downloaded from the database and the ligand was
removed from the fragment. The resulting macromolecule was setup
for docking as follows: polar hydrogens were added using the
PROTONATE utility. Solvation parameters were added to the final
protein file using the ADDSOL utility.
The grid maps representing the protein in the actual docking
process were calculated with AutoGrid. The grids (one for each
atom type in the ligand, plus one for electrostatic interactions)
were chosen to be sufficiently large to include not only the active
site but also significant portions of the surrounding surface. The
points of the grids were thus 60 × 60 × 60, with a grid spacing
of 0.375 Å.
position drug molecules on DNA for other reaction responsible
of the cytotoxic and antitumor activity of these compounds.
Conclusions
In conclusion, this class of compounds combines two of the
main features of DNA-interactive drugs and a very tight DNA-
binding capability has to be expected, as suggested by our
docking studies. The molecular modeling results correlate well
with the experimental activity data and the choice of suitable
chains can be theoretically supported: a more negative ∆Gbinding
value is mirrored by a higher inhibitory activity. The class of
4-substituted indolo[3,2-e][1,2,3]triazolo[1,5-a]pyrimidines stud-
ied herein represents an improvement of the original lead
Docking. Docking of the ligands into receptor was carried out
using AutoDock (version 3.0.5) set of programs.23 It was carried

Indolo[3,2-e][1,2,3]triazolo[1,5-a]pyrimidine
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 7 2043
Table 6. Overview of the Results (pTGI and pLC50) of the In Vitro Antitumor Screening for Indolo-triazolo-pyrimidinesa
c
# cell lines
investigated
# cell lines giving
positive pTGI
pTGIb
range
# cell lines
investigated
# cell lines giving
positive pLC50
pLC50
cmpd
MG_MIDd
range
MG_MID
1eK
4
1eL
1eJ
56
51
56
56
19
21
22
52
4.97-4.02
5.29-4.01
5.45-4.31
5.26-4.33
4.09
4.18
4.46
4.67
52
56
56
7
5
38
5.07-4.01
4.77-4.32
4.57-4.31
4.03
4.32
4.39
a Data obtained from the NCI’s in vitro disease-oriented human tumor cells screen. b pTGI is the -log of the molar concentration giving total growth
inhibition. c pLC50 is the -log of the molar concentration leading to 50% net cell death. d MG_MID ) mean graph midpoint ) arithmetical mean value for
all tested cancer cell lines. If the indicated effect was not attainable within the used concentration interval, the highest tested concentration was used for the
calculation.
out using the empirical free energy function and the Lamarckian
genetic algorithm,14 applying a standard protocol, with an initial
population of 100 randomly placed individuals, a maximum number
of 1.5 × 106 energy evaluations, a mutation rate of 0.02, a crossover
rate of 0.80, and an elitism value of 1. Proportional selection was
used, where the average of the worst energy was calculated over a
window of the previous 10 generations. For the local search, the
so-called pseudo-Solis and Wets algorithm24 was applied using a
maximum of 300 iterations per local search. The probability of
performing local search on an individual in the population was 0.06,
and the maximum number of consecutive successes or failures
before doubling or halving the local search step size was 4. A total
of 10 independent docking runs were carried out for each ligand.
At the end of each AutoDock execution, in which more than one
run was performed, the program outputs a list of clusters and their
energies. The clustering of docked conformations is determined by
the tolerance specified in Å. In our cases, the results differing by
less than 1.0 Å in positional root-mean-square deviation (rmsd)
were clustered together and represented by the result with the most
favorable free energy of binding. The best representative from each
cluster (the one with the lowest energy) is written out in PDBQ
format at the end of the log file. These structures can be read into
any appropriate molecular modeling system. The table of ranked
clusters shows the final docked energy for each conformation and
the rms difference between the lowest energy member of the group
and every other member. The rms for the lowest member of the
group is by definition zero. After this table, the structures are output
in PDBQ format. Each conformation has a set of remark records,
one of which describes the rms difference between itself and the
coordinates specified in the original input PDBQ file.
Chemistry. All melting points were taken on a Buchi-Tottoli
capillary apparatus and are uncorrected; IR spectra were determined
in bromoform with a Jasco FT/IR 5300 spectrophotometer; 1H and
13C NMR spectra were measured in DMSO-d6 solution (TMS as
internal reference) at 200 and 50.3 MHz, respectively, using a
Bruker AC-E series 200 MHz spectrometer. 13C NMR spectra are
reported, indicating the multiplicity, (s, singlet; d, doublet; t, triplet;
q, quartet) assigned by DEPT135 experiments. Column chroma-
tography was performed with Merck silica gel 230–400 mesh
ASTM or with a Biotage FLASH40i chromatography module
(prepacked cartridge system). Microanalyses were in agreement with
theoretical values (0.4%.
4-[3-(Dibutylamino)propyl]-10-methyl-3-phenyl-4H-indolo[3,2-
e][1,2,3]triazolo[1,5-a]pyrimidin-5(10H)-one (1eK). A stirred
suspension of 2 (0.5 g, 1.28 mmol) in dibutylamine (20 mL) was
heated at reflux for 5 h. After cooling water (10 mL) was added
and the mixture was carefully adjusted to pH ) 7 with 6 M
hydrochloric acid, extracted with dichloromethane, dried over
sodium sulfate, and evaporated to dryness. The residue was purified
by flash chromatography using ethyl acetate as eluant to afford
compound 1eK as a white solid (0.43 g, 0.89 mmol). Yield 70%,
1
mp 205 °C. IR: 1656 (CdO) cm-1. H NMR ppm: 8.48 (d, J )
7.4 Hz, 2H, H-2′ and H-6′); 8.07 (d, J ) 7.4 Hz, 1H, H-6);
7.60–7.30 (m, 6H, H-7, H-8, H-9, H-3′, H-4′ and H-5′); 5.03 (t, J
) 6.6 Hz, 2H, CONCH2CH2CH2N); 3.90 (s, 3H, NCH3); 2.39 (t,
J ) 7.4 Hz, 2H, NCH2CH2CH2N); 2.27–2.24 (m, 4H, 2 ×
NCH2CH2CH2CH3); 1.86–1.83 (m, 2H, CONCH2CH2CH2N);
1.23–1.20 (m, 8H, 2 × NCH2CH2CH2CH3); 0.80–0.75 (m, 6H, 2
× NCH2CH2CH2CH3). 13C NMR ppm: 154.6 (s); 151.6 (s); 144.8
(s); 138.3 (s); 130.9 (s); 129.7 (d); 128.9 (d); 128.6 (s); 126.3 (d);
124.7 (d); 121.4 (d); 121.1 (s); 120.3 (d); 109.8 (d); 94.6 (s); 53.7
(t); 52.4 (t); 49.6 (t); 28.0 (q); 26.9 (t); 19.7 (t); 13.6 (q).
Ethyl 4-(10-Methyl-3-phenyl-5-oxo-5,10-dihydro-4H-indolo[3,2-
e][1,2,3]triazolo[1,5-a]pyrimidin-4-yl)butanoate (1eN). To a stirred
suspension of 1e9 (0.5 g, 1.6 mmol) in dry acetonitrile, (20 mL)
potassium carbonate (1.09 g, 7.9 mmol) and ethyl 4-bromobutyrate
(1.13 mL, 7.9 mmol) were added. The mixture was heated at reflux
for 5 h, cooled to room temperature, and evaporated under reduced
pressure. The residue was purified by column chromatography using
dichlorometane as eluant to afford compound 1eN as a white solid
(0.48 g, 1.12 mmol.). Yield 70%, mp 128.9 °C. IR: 1729 (CdO),
1691 (CdO) cm-1. 1H NMR ppm: 8.45 (d, J ) 7.4 Hz, 2H, H-2′
and H-6′); 8.05 (d, J ) 7.4 Hz, 1H, H-6); 7.63–7.29 (m, 6H, H-7,
H-8, H-9, H-3′, H-4′ and H-5′); 5.03 (t, J ) 6.6 Hz, 2H, NCH2);
3.97–3.88 (m, 5H, NCH3 and OCH2CH3); 2.38 (t, J ) 7.4 Hz, 2H,
CH2CO); 2.07–1.99 (m, 2H, CH2CH2CH2); 1.06 (t, J ) 7.3 Hz,
3H, OCH2CH3). 13C NMR ppm: 171.9 (s); 154.4 (s); 151.3 (s);
144.5 (s); 138.1 (s); 136.1 (s); 129.4 (d); 128.7 (d); 128.6 (s); 126.1
(d); 124.4 (d); 121.2 (d); 121.1 (s); 120.2 (d); 109.5 (d); 94.6 (s);
59.8 (t); 55.0 (t); 30.4 (t); 27.9 (q); 23.5 (t); 13.8 (q).
4-(10-Methyl-3-phenyl-5-oxo-5,10-dihydro-4H-indolo[3,2-
e][1,2,3]triazolo[1,5-a]pyrimidin-4-yl)butanoic Acid (1eI). To a
stirred suspension of 1eN (0.5 g, 1.16 mmol) in dry ethanol (10
mL), a solution of sodium hydroxide (0.23 g, 5.8 mmol) in water
(10 mL) was added dropwise. The mixture was heated at reflux
for 1 h until a clear solution was obtained. The ethanol was
evaporated under reduced pressure and the aqueous layer was then
carefully adjusted to pH )1 with 6 M hydrochloric acid. The
resulting solid was filtered off, air-dried, and recrystallized from
ethanol to afford compound 1eI as a colorless solid (0.46 g, 1.16
mmol). Yield 100%, mp 219.3 °C. IR: 3519–2900 (OH), 1716
4-(3-Chloropropyl)-10-methyl-3-phenyl-4H-indolo[3,2-e][1,2,3]-
triazolo[1,5-a]pyrimidin-5(10H)-one (2). To a stirred suspension
of 1e9 (0.5 g, 1.6 mmol) in dry acetonitrile (20 mL), potassium
carbonate (0.66 g, 4.8 mmol) and 1-bromo-3-chloropropane (0.47
mL, 4.8 mmol) were added. The mixture was heated at reflux for
5 h, cooled to room temperature, and evaporated under reduced
pressure. The residue was purified by column chromatography using
dichloromethane as eluant to afford compound 2 as a white solid
1
(CdO), 1681 (CdO) cm-1. H NMR ppm: 8.47 (d, J ) 7.3 Hz,
(0.44 g, 1.12 mmol). Yield 70%, mp 202 °C. IR: 1687 (CdO) cm-1
.
2H, H-2′ and H-6′); 8.07 (d, J ) 7.3 Hz, 1H, H-6); 7.66–7.28 (m,
7H, H-7, H-8, H-9, H-3′, H-4′, H-5′ and OH); 5.02 (t, J ) 5.9 Hz,
2H, NCH2); 3.89 (s, 3H, NCH3); 2.32 (t, J ) 7.4 Hz, 2H, CH2CO);
2.08–1.94 (m, 2H, CH2CH2CH2). 13C NMR ppm: 173.6 (s); 154.5
(s); 151.4 (s); 144.6 (s); 138.2 (s); 136.2 (s); 129.5 (d); 128.9 (d);
128.7 (s); 126.2 (d); 124.6 (d); 121.3 (d); 121.2 (s); 120.3 (d); 109.6
(d); 94.7 (s); 55.2 (t); 30.6 (t); 27.9 (q); 23.6 (t).
1H NMR ppm: 8.28 (d, J ) 7.4 Hz, 2H, H-2′ and H-6′); 8.08 (d,
J ) 7.4 Hz, 1H, H-6); 7.62–7.31 (m, 6H, H-7, H-8, H-9, H-3′,
H-4′ and H-5′); 5.11 (t, J ) 7.3 Hz, 2H, NCH2); 3.92 (s, 3H, NCH3);
3.72 (t, J ) 5.9 Hz, 2H, CH2Cl); 2.30–2.24 (m, 2H, CH2CH2CH2).
13C NMR ppm: 156.0 (s); 151.2 (s); 144.5 (s); 138.1 (s); 135.1
(s); 129.5 (d); 128.8 (d); 128.6 (s); 126.3 (d); 124.5 (d); 121.3 (d);
121.1 (s); 120.3 (d); 109.6 (d); 94.6 (s); 53.2 (t); 42.1 (t); 30.8 (t);
27.9 (q).
Ethyl N-[4-(10-Methyl-3-phenyl-5-oxo-5,10-dihydro-4H-in-
dolo[3,2-e][1,2,3]triazolo[1,5-a]pyrimidin-4-yl)butanoyl]glyci-

2044 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 7
Lauria et al.
nate (1eL). To a stirred suspension of carboxylic acid 1eI (0.5 g,
1.25 mmol) in dry dichloromethane (10 mL), cooled to 0 °C, DCC
(0.26 g, 1.25 mmol) and triethylamine (0.17 mL, 1.25 mmol) were
added. After 2 h, glycine ethyl ester hydrochloride (0.17 g, 1.25
mmol) was added and the mixture was stirred for further 5 h at
room temperature. Evaporation of the solvent under reduced
pressure gave a solid that was purified by column chromatography.
Using dichloromethane as eluant, compound 5 (<5% yield) was
isolated. Elution with dichloromethane/ethyl acetate 9:1 gave
compound 1eL as a white solid (0.30 g, 0.63 mmol). Yield 50%,
mp 217 °C. IR: 3322 (NH), 1752 (CdO), 1693 (CdO), 1650
Hz, 2H, NCH2); 4.43 (m, 1H, CH); 3.89 (s, 3H, NCH3); 3.61 (d, J
) 5.1 Hz, 2H, NHCH2CO); 3.56 (s, 3H, OCH3); 2.86 (d, J ) 5.1
Hz, 2H, CHCH2); 2.22 (t, J ) 7.3 Hz, 2H, CH2CO); 2.02–1.96 (m,
2H, CH2CH2CH2). 13C NMR ppm: 171.5 (s); 171.3 (s); 169.0 (s);
154.0 (s); 151.4 (s); 144.5 (s); 138.1 (s); 136.2 (s); 134.8 (d); 131.3
(s); 129.4 (d); 128.8 (d); 128.6 (s); 126.2 (d); 124.5 (d); 121.2 (d);
121.1 (s); 120.2 (d); 116.6 (d); 109.5 (d); 94.7 (s); 55.4 (t); 52.3
(d); 51.8 (q); 41.6 (t); 31.8 (t); 27.9 (q); 27.4 (t); 24.1 (t).
Methyl N-[4-(10-Methyl-3-phenyl-5-oxo-5,10-dihydro-4H-in-
dolo[3,2-e][1,2,3]triazolo[1,5-a]pyrimidin-4-yl)butanoyl]histidi-
nate (1eQ) and N-Cyclohexyl-N-[(cyclohexylamino)carbonyl]4-
(10-methyl-3-phenyl-5-oxo-5,10-dihydro-4H-indolo[3,2-
e][1,2,3]triazolo[1,5-a]pyrimidin-4-yl)butanamide (5). To a stirred
suspension of carboxylic acid 1eI (0.5 g, 1.25 mmol) in dry
dichloromethane (10 mL), cooled to 0 °C, DCC (0.26 g, 1.25 mmol)
and triethylamine (0.17 mL, 1.25 mmol) were added. After 2 h,
L-histidine methyl ester dihydrochloride (0.3 g, 1.25 mmol) was
added and the mixture was stirred for an additional 24 h at room
temperature. Evaporation of the solvent under reduced pressure gave
a solid that was purified by column chromatography. Using
dichloromethane as eluant compound 5 (0.16 g, 0.27 mmol) was
isolated. Yield 22%, mp 215.9 °C. IR: 3326 (NH), 1677 (CdO),
1652 (CdO) cm-1. 1H NMR ppm: 8.48 (d, J ) 7.4 Hz, 2H, H-2′
and H-6′); 8.08 (d, J ) 7.4 Hz, 1H, H-6); 7.61–7.29 (m, 6H, H-7,
H-8, H-9, H-3′, H-4′ and H-5′); 5.58 (d, J ) 7.3 Hz, 1H, NH);
5.05 (t, J ) 6.6 Hz, 2H, NCH2); 3.92 (s, 3H, NCH3); 2.31 (t, J )
7.4 Hz, 2H, CH2CO); 2.05–2.02 (m, 2H, CH2CH2CH2); 1.70–1.06
(m, 22H, 2 × Cy).
1
(CdO) cm-1. H NMR ppm: 8.47 (d, J ) 7.3 Hz, 2H, H-2′ and
H-6′); 8.15 (t, J ) 5.1 Hz, 1H, NHCH2); 8.07 (d, J ) 7.3 Hz, 1H,
H-6); 7.64–7.26 (m, 6H, H-7, H-8, H-9, H-3′, H-4′ and H-5′); 5.01
(t, J ) 6.6 Hz, 2H, NCH2); 4.07 (q, J ) 7.3 Hz, 2H, OCH2CH3);
3.89 (s, 3H, NCH3); 3.61 (d, J ) 5.1 Hz, 2H, NHCH2CO); 2.22 (t,
J ) 7.3 Hz, 2H, CH2CO); 2.02–1.96 (m, 2H, CH2CH2CH2); 1.06
(t, J ) 7.3 Hz, 3H, OCH2CH3). 13C NMR ppm: 173.6 (s); 171.2
(s); 154.5 (s); 151.4 (s); 144.6 (s); 138.2 (s); 136.2 (s); 129.5 (d);
128.9 (d); 128.7 (s); 126.2 (d); 124.6 (d); 121.3 (d); 121.2 (s); 120.3
(d); 109.6 (d); 94.7 (s); 55.2 (t); 55.0 (t); 41.6 (t); 30.6 (t); 27.9
(q); 23.6 (t); 13.8 (t).
N-[4-(10-Methyl-3-phenyl-5-oxo-5,10-dihydro-4H-indolo[3,2-
e][1,2,3]triazolo[1,5-a]pyrimidin-4-yl)butanoyl]glycine (3). To a
stirred suspension of 1eL (0.5 g, 1.02 mmol) in dry ethanol (10
mL), a solution of sodium hydroxide (0.2 g, 5.1 mmol) in water
(10 mL) was added dropwise. The mixture was incubated at 50 °C
for 3 h until a clear solution was obtained. The ethanol was
evaporated under reduced pressure and the aqueous layer was then
carefully adjusted to pH ) 1 with 6 M hydrochloric acid. The
resulting solid was filtered off, air-dried, and recrystallized from
ethanol to afford compound 3 as a white solid (0.47 g, 1.02 mmol).
Yield 100%, mp 215 °C. IR: 3446–2900 (OH and NH), 1731
Further elution with dichloromethane/methanol 9:1 gave com-
pound 1eQ (0.35 g, 0.63 mmol). Yield 50%, mp 187 °C. IR: 3342
1
(NH), 1737 (CdO), 1691 (CdO), 1656 (CdO) cm-1. H NMR
ppm: 8.46 (d, J ) 7.4 Hz, 2H, H-2′ and H-6′); 8.29 (d, J ) 7.3 Hz,
1H, NH); 8.05 (d, J ) 7.4 Hz, 1H, H-6); 7.60–7.29 (m, 7H, H-7,
H-8, H-9, H-3′, H-4′, H-5′ and H-2′′); 6.77 (s, 1H, H-5′′); 4.96 (t,
J ) 6.6 Hz, 2H, NCH2); 4.44 (m, 1H, CH); 3.88 (s, 3H, NCH3);
3.56 (s, 3H, OCH3); 2.83 (d, J ) 7.4 Hz, 2H, CHCH2); 2.20 (t, J
) 7.4 Hz, 2H, CH2CO); 2.06–1.96 (m, 2H, CH2CH2CH2). 13C NMR
ppm: 172.1 (s); 171.2 (s); 154.4 (s); 151.4 (s); 144.6 (s); 138.1 (s);
136.2 (s); 134.9 (d); 133.1 (s); 129.5 (d); 128.9 (d); 128.7 (s); 126.2
(d); 124.5 (d); 121.3 (d); 121.1 (s); 120.3 (d); 116.6 (d); 109.6 (d);
94.7 (s); 55.3 (t); 52.4 (d); 51.8 (q); 31.7 (t); 28.9 (t); 27.9 (q);
24.1 (t).
1
(CdO), 1666 (CdO) cm-1. H NMR ppm: 8.47 (d, J ) 7.3 Hz,
2H, H-2′ and H-6′); 8.15 (t, 1H, J ) 5.1 Hz, NH); 8.07 (d, J ) 7.3
Hz, 1H, H-6); 7.66–7.28 (m, 6H, H-7, H-8, H-9, H-3′, H-4′ and
H-5′); 5.02 (t, J ) 6.6 Hz, 2H, NCH2CH2CH2) 3.89 (s, 3H, NCH3);
3.61 (d, 2H, J ) 5.1 Hz, NHCH2CO); 2.32 (t, J ) 7.4 Hz, 2H,
CH2CH2CH2CO); 2.04–1.99 (m, 2H, CH2CH2CH2). 13C NMR ppm:
173.6 (s); 171.2 (s); 154.5 (s); 151.4 (s); 144.6 (s); 138.2 (s); 136.2
(s); 129.5 (d); 128.9 (d); 128.7 (s); 126.2 (d); 124.6 (d); 121.3 (d);
121.2 (s); 120.3 (d); 109.6 (d); 94.7 (s); 55.2 (t); 41.6 (t); 30.6 (t);
27.9 (q); 23.6 (t).
Synthesis of 1eQ and 1eR by Using DCI. To a stirred solution
of the acid, 1eI or 3 (1.25 mmol), in dry dioxane (10 mL), cooled
to 0 °C, DCI (0.3 g, 1.87 mmol) was added. After 5 h, L-histidine
methyl ester dihydrochloride (0.3 g, 1.25 mmol) and triethylamine
(0.26 mL, 1.87 mmol) were added and the mixture was incubated
at 80 °C for 8 h. Evaporation of the solvent under reduced pressure
gave a solid that was purified by column chromatography using
dichloromethane/methanol 9:1 as eluant to afford compounds 1eQ
or 1eR (yields 80 and 70%, respectively).
Methyl N-[4-(10-Methyl-3-phenyl-5-oxo-5,10-dihydro-4H-
indolo[3,2-e][1,2,3]triazolo[1,5-a]pyrimidin-4-yl)butanoyl]gly-
cinhistidinate (1eR) and N-(2-{Cyclohexyl[(cyclohexylamino)-
carbonyl]amino}-2-oxoethyl)-4-(10-methyl-3-phenyl-5-oxo-5,
10-dihydro-4H-indolo[3,2-e][1,2,3]triazolo[1,5-a]pyrimidin-4-
yl)butanamide (4). To a stirred suspension of 3 (0.5 g, 1.09 mmol)
in dry dichloromethane (10 mL), cooled to 0 °C, DCC (0.22 g,
1.09 mmol) and triethylamine (0.15 mL, 1.09 mmol) were added.
After 2 h, L-histidine methyl ester dihydrochloride (0.26 g, 1.09
mmol) was added, and the mixture was stirred for further 24 h at
room temperature. Evaporation of the solvent under reduced
pressure gave a solid that was purified by column chromatography.
Using dichloromethane as eluant compound 4 as a colorless solid
(0.16 g, 0.24 mmol) was isolated. Yield 22%, mp 152 °C. IR: 3324
(NH), 2929 (NH), 1691 (CdO), 1652 (CdO) cm-1. 1H NMR ppm:
8.54 (d, J ) 7.4 Hz, 2H, H-2′ and H-6′); 8.32 (t, J ) 5.1 Hz, 1H,
NHCH2); 8.09 (d, J ) 7.4 Hz, 1H, H-6); 7.60–7.34 (m, 6H, H-7,
H-8, H-9, H-3′, H-4′ and H-5′); 5.59 (d, J ) 7.3 Hz, 1H, NH);
5.05 (t, J ) 6.6 Hz, 2H, NCH2); 3.95 (s, 3H, NCH3); 3.82 (d, J )
5.1 Hz, 2H, NHCH2CO); 2.24 (t, J ) 7.4 Hz, 2H, CH2CO);
2.04–1.97 (m, 2H, CH2CH2CH2); 1.70–1.24 (m, 22H, 2 × Cy).
Further elution with dichloromethane/methanol 9:1 gave com-
pound 1eR as a colorless solid (0.33 g, 0.54 mmol). Yield 50%,
mp 147 °C. IR: 3313 (NH), 2948 (NH), 1737 (CdO), 1685 (CdO)
N-[2-(1H-Imidazol-4-yl)ethyl]-4-(10-methyl-3-phenyl-5-oxo-
5,10-dihydro-4H-indolo[3,2-e][1,2,3]triazolo[1,5-a]pyrimidin-4-
yl)butanamide (1eJ). To a stirred solution of the acid 1eI (0.5 g,
1.25 mmol) in dry dioxane (10 mL), cooled to 0 °C, DCI (0.3 g,
1.87 mmol) was added. After 5 h, histamine-free base (0.2 g, 1.87
mmol) and triethylamine (0.26 mL, 1.87 mmol) were added and
the mixture was incubated at 50 °C for 1 h. Evaporation of the
solvent under reduced pressure gave a solid that was purified by
column chromatography using dichloromethane/methanol 9:1 as
eluant to afford compound 1eJ (0.35 g, 0.7 mmol). Yield 56%, mp
207 °C. IR: 3360 (NH), 1716 (CdO), 1681 (CdO) cm-1. 1H NMR
ppm: 8.50 (d, J ) 7.3 Hz, 2H, H-2′ and H-6′); 8.09 (d, J ) 7.3 Hz,
1H, H-6); 7.65 (t, J ) 7.3 Hz, 1H, NHCH2); 7.52–7.29 (m, 7H,
H-7, H-8, H-9, H-3′, H-4′, H-5′ and H-2”); 6.75 (s, 1H, H-5′′);
5.02 (t, J ) 6.6 Hz, 2H, NCH2); 3.93 (s, NCH3); 3.36–3.17 (m,
4H, NHCH2CH2 and NHCH2CH2); 2.20 (t, J ) 7.4 Hz, 2H,
CH2CO); 2.07–1.98 (m, 2H, CH2CH2CH2). 13C NMR ppm: 170.9
(s); 154.6 (s); 151.5 (s); 145.9 (s); 138.2 (s); 136.2 (s); 134.5 (d);
133.1 (s); 129.4 (d); 128.9 (d); 128.7 (s); 126.1 (d); 124.6 (d); 121.4
1
cm-1. H NMR ppm: 8.47 (d, J ) 7.3 Hz, 2H, H-2′ and H-6′);
8.28 (d, J ) 7.4 Hz, 1H, NHCH); 8.15 (t, J ) 5.1 Hz, 1H, NHCH2);
8.07 (d, J ) 7.3 Hz, 1H, H-6); 7.64–7.26 (m, 7H, H-7, H-8, H-9,
H-3′, H-4′, H-5′ and H-2′′); 6.79 (s, 1H, H-5′′); 5.01 (t, J ) 6.6

Indolo[3,2-e][1,2,3]triazolo[1,5-a]pyrimidine
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 7 2045
(d); 121.2 (s); 120.3 (d); 116.6 (d); 109.7 (d); 94.6 (s); 55.3 (t);
31.9 (t); 28.3 (t); 27.9 (q); 26.9 (t); 24.1 (t).
Cancer Institute, especially Dr. V. L. Narayanan, Dr. R.
Shoemaker, and their teams, for the antitumor tests reported in
this paper and for the helpful discussions.
Biology. Methodology of the In Vitro Cancer Screen. The
human tumor cell lines of the cancer screening panel are grown in
RPMI 1640 medium containing 5% fetal bovine serum and 2 mM
L-glutamine. For a typical screening experiment, cells are inoculated
into 96-well microtiter plates in 100 µL at plating densities ranging
from 5000 to 40000 cells/well, depending on the doubling time of
individual cell lines. After cell inoculation, the microtiter plates
are incubated at 37 °C, 5% CO2, 95% air, and 100% relative
humidity for 24 h prior to addition of experimental drugs. After
24 h, two plates of each cell line are fixed in situ with TCA, to
represent a measurement of the cell population for each cell line at
the time of drug addition (Tz). Experimental drugs are solubilized
in DMSO at 400-fold the desired final maximum test concentration
and stored frozen prior to use. At the time of drug addition, an
aliquot of frozen concentrate is thawed and diluted to twice the
desired final maximum test concentration with complete medium
containing 50 µg/mL gentamicin. Additional four, 10-fold, or ½
log serial dilutions are made to provide a total of five drug
concentrations plus control. Aliquots of 100 µL of these different
drug dilutions are added to the appropriate microtiter wells already
containing 100 µL of medium, resulting in the required final drug
concentrations.
Following drug addition, the plates are incubated for an additional
48 h at 37 °C, 5% CO2, 95% air, and 100% relative humidity. For
adherent cells, the assay is terminated by the addition of cold TCA.
Cells are fixed in situ by the gentle addition of 50 µL of cold 50%
(w/v) TCA (final concentration, 10% TCA) and incubated for 60
min at 4 °C. The supernatant is discarded, and the plates are washed
five times with tap water and air-dried. Sulforhodamine B (SRB)
solution (100 µL) at 0.4% (w/v) in 1% acetic acid is added to each
well, and plates are incubated for 10 min at room temperature. After
staining, unbound dye is removed by washing five times with 1%
acetic acid and the plates are air-dried. Bound stain is subsequently
solubilized with 10 mM trizma base, and the absorbance is read
on an automated plate reader at a wavelength of 515 nm. For
suspension cells, the methodology is the same except that the assay
is terminated by fixing settled cells at the bottom of the wells by
gently adding 50 µL of 80% TCA (final concentration, 16% TCA).
Using the seven absorbance measurements [time zero, (Tz), control
growth, (C), and test growth in the presence of drug at the five
concentration levels (Ti)], the percentage growth is calculated at
each of the drug concentrations levels. Percentage growth inhibition
is calculated as
Supporting Information Available: Elemental analyses for all
compounds and table with full biological data. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) (a) Aivasashvilli, V. A.; Beabealashvilli, R. S. Sequence-specific
inhibition of RNA elongation by actinomycin D. FEBS Lett. 1983,
160, 124–128. (b) Lane, M. J.; Dabrowiak, J. C.; Vournakis, J.
Sequence specificity of actinomycin D and Netropsin binding to
pBR322 DNA analyzed by protection from DNase I. Proc. Natl. Acad.
Sci. U.S.A. 1983, 80, 3260–3264. (c) Fox, K. R.; Waring, M. DNA
structural variations produced by actinomycin and distamycin as
revealed by DNAase I footprinting. Nucleic Acids Res. 1984, 12, 9271–
9285. (d) Phillips, D. R.; Crothers, D. M. Biochemistry Kinetics and
sequence specificity of drug-DNA interactions: an in vitro transcription
assay. Biochemistry 1986, 25, 7355–7362. (e) Goodisman, J.; Rehfuss,
R.; Ward, B.; Dabrowiak, J. C. Site-specific binding constants for
actinomycin D on DNA determined from footprinting data. Biochem-
istry 1992, 31, 1046–1058.
(2) (a) Patel, D. J. Peptide antibiotic-oligonucleotide interactions. Nuclear
magnetic resonance investigations of complex formation between
actinomycin D and d-ApTpGpCpApT in aqueous solution. Biochem-
istry 1974, 13, 2396–2402. (b) Liu, X.; Chen, H.; Patel, D. J. Solution
structure of actinomycin-DNA complexes: Drug intercalation at
isolated G-C sites. J. Biomol. NMR 1991, 1, 323–347. (c) Brown,
D. R.; Kurz, M.; Kearns, D. R.; Hsu, V. L. Formation of multiple
complexes between actinomycin D and a DNA hairpin: Structural
characterization by multinuclear NMR. Biochemistry 1994, 33, 651–
664.
(3) (a) Sobell, H. M.; Jain, S. C.; Sakore, T. D.; Nordman, C. E.
Stereochemistry of actinomycin-DNA binding. Nature 1971, 231,
200–205. (b) Sobell, H. M.; Jain, S. C. Stereochemistry of actinomycin
binding to DNA. II. Detailed molecular model of actinomycin-DNA
complex and its implications. J. Mol. Biol. 1972, 68, 21–34. (c)
Kamitori, S.; Takusagawa, F. Crystal structure of the 2:1 complex
between d(GAAGCTTC) and the anticancer drug actinomycin D. J.
Mol. Biol. 1992, 225, 446–456. (d) Kamitori, S.; Takusagawa, F.
Multiple binding modes of anticancer drug actinomycin D: X-ray, molecular
modeling, and spectroscopic studies of d(GAAGCTTC)2-actinomycin
D complexes and its host DNA. J. Am. Chem. Soc. 1994, 116, 4154–
4165.
(4) (a) Rill, R. L.; Marsch, G. A.; Graves, D. E. 7-Azido-actinomycin D:
Photoaffinity probe of the sequence specificity of DNA binding by
actinomycin D. J. Biomol. Struct. Dyn. 1989, 7, 591–605.
(5) Lian, C. Y.; Robinson, H.; Wang, A. H. J. Structure of actinomycin
D bound with (GAAGCTTC)2 and (GATGCTTC)2 and its binding to
the (CAG)n:(CTG)n triplet sequence by NMR analysis. J. Am. Chem.
Soc. 1996, 118, 8791–8801.
[(Ti − Tz) ⁄ (C − Tz)] × 100 for concentrations for which Ti ≥ Tz
[(Ti − Tz) ⁄ Tz] × 100 for concentrations for which Ti < Tz
(6) Bailey, S. A.; Graves, D. E.; Rill, R.; Marsch, G. Influence of DNA
base sequence on the binding energetics of actinomycin D. Biochem-
istry 1993, 32, 5881–5887.
Three dose-response parameters are calculated for each experi-
mental agent. Growth inhibition of 50% (GI50) is calculated from
[(Ti - Tz)/(C - Tz)] × 100 ) 50, which is the drug concentration,
resulting in a 50% reduction in the net protein increase (as measured
by SRB staining) in control cells during the drug incubation. The
drug concentration resulting in total growth inhibition (TGI) is
calculated from Ti ) Tz. The LC50 (concentration of drug resulting
in a 50% reduction in the measured protein at the end of the drug
treatment as compared to that at the beginning) indicating a net
loss of cells following treatment is calculated from [(Ti - Tz)/Tz]
× 100 ) -50. Values are calculated for each of these three
parameters if the level of activity is reached; however, if the effect
is not reached or is exceeded, the value for that parameter is
expressed as greater or less than the maximum or minimum
concentration tested.
(7) Lauria, A.; Diana, P.; Barraja, P.; Almerico, A. M.; Cirrincione, G.;
Dattolo, G. Pyrrolo[3,4-e][1,2,3]triazolo[1,5-a]pyrimidine and pyr-
rolo[3,4-d][1,2,3]-triazolo[1,5-a]pyrimidine. New tricyclic ring systems
of biological interest. J. Heterocycl. Chem. 2000, 37, 747–750.
(8) Lauria, A.; Diana, P.; Barraja, P.; Montalbano, A.; Cirrincione, G.;
Dattolo, G.; Almerico, A. M. New tricyclic systems of biological
interest. Annelated 1,2,3-triazolo[1,5-a]pyrimidines through domino
reaction of 3-azidopyrroles and methylene active nitriles. Tetrahedron
2002, 58, 9723–9727.
(9) Lauria, A.; Patella, C.; Diana, P.; Barraja, P.; Montalbano, A.;
Cirrincione, G.; Dattolo, G.; Almerico, A. M. A new tetracyclic ring
system of biological interest. Indolo[3,2-e][1,2,3]triazolo[1,5-a]pyri-
midines through domino reaction of 2-azidoindole. Heterocycles 2003,
60, 2669–2675.
(10) Lauria, A.; Patella, C.; Diana, P.; Barraja, P.; Montalbano, A.;
Cirrincione, G.; Dattolo, G.; Almerico, A. M. A synthetic approach
to new polycyclic ring system of biological interest through domino
reaction: indolo[2,3-e][1,2,3]triazolo[1,5-a]pyrimidine. Tetrahedron
Lett. 2006, 47, 2187–2190.
(11) Web address: http://www.ccdc.cam.ac.uk.
(12) Web address: http://www.rcsb.org/pdb.
Adriamycin was tested with each screening run as a quality
control measure. This is done to establish the ability of the screen
to generate information relevant to mechanism of growth inhibition/
cell killing. It is not a reference compound that has unique
importance for the compounds.
(13) Web address: http://www.scripps.edu/pub/olson-web/doc/autodock.
(14) Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart, W. E.;
Belew, R. K.; Olson, A. J. Automated docking using Lamarckian
genetic algorithm and an empirical binding free energy function.
J. Comput. Chem. 1998, 19, 1639–1662.
Acknowledgment. This work was financially supported by
Ministero dell’Università e della Ricerca. We thank the National

2046 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 7
Lauria et al.
(15) Lauria, A.; Patella, C.; Diana, P.; Barraja, P.; Montalbano, A.;
Cirrincione, G.; Dattolo, G.; Almerico, A. M. MADoSPRO: A new
approach to molecular modeling studies on a series of DNA minor
groove binders. QSAR Comb. Sci. 2006, 25, 252–262.
(16) (a) See, for example, Doleschall, G.; Lempert, K. Mechanism of
carboxyl condensations by carbodiimides. Tetrahedron Lett. 1963,
1195–1199. (b) Hegarty, A. F.; McCormack, M. T.; Ferguson, G.;
Roberts, P. J. Isolation and reactivity of a model for the carbodiimide-
carboxylic acid adduct. O-Benzoyl-N,N-dimethyl-N′-(N-methyl-2,4-
dinitroanilino)isourea. J. Am. Chem. Soc. 1977, 99, 2015–2016.
(17) Mikolajczyk, M.; Kielbasinski, P. Recent developments in the car-
bodimide chemistry. Tetrahedron 1981, 37, 233–284.
(18) (a) Grever, M. R.; Sherpartz, S. A.; Chabner, B. A. The National
Cancer Institute: Cancer drug discovery and development program.
Semin. Oncol. 1992, 19, 622–638. (b) Monks, A. P.; Scudiero, D. A.;
Skehan, P.; Shoemaker, R.; Paull, K. D.; Vistica, D.; Hose, C.;
Langley, J.; Croniste, P.; Vaigro-Woiff, A.; Gray-Goodrich, M.;
Campbell, H.; Mayo, J.; Boyd, M. R. Feasibility of a high-flux
anticancer drug screen utilizing a derived panel of human tumor cell
lines in culture. J. Natl. Cancer Inst. 1991, 83, 757–766. (c) Weinstein,
J. N.; Meyers, T. G.; O’Connor, P. M.; Friend, S. H.; Fornace, A. J.,
Jr.; Kohn, K. W.; Fojo, T.; Bates, S. E.; Rubinstein, L, V.; Anderson,
N. L.; Boulamwini, J. K.; van Osdol, W. W.; Monks, A. P.; Scudiero,
D. A.; Sausville, E. A.; Zaharevitz, D. W.; Bunow, B.; Viswanadhan,
V. N.; Johnson, G. S.; Wittes, R. E.; Paull, K. D. An information-
intensive approach to the molecular pharmacology of cancer. Science
1997, 275, 343–349.
(19) Data available on line at http://dtp.nci.nih.gov/dtpstandard/cancer-
screeningdata/index.jsp (code: NSC3053).
(20) Paull, K. D.; Shoemaker, R. H.; Hodes, L.; Monks, A.; Scudiero, D. A.;
Rubistein, L.; Plowman, J.; Boyd, M. R., J. Display and analysis of
patterns of differential activity of drugs against human tumor cell lines:
Development of mean graph and COMPARE algorithm. Natl. Cancer
Inst. 1989, 81, 1088–1092.
(21) Available online at http://spheroid.ncifcrf.gov/default.html and devel-
oped according to the following: Rabow, A. A.; Shoemaker, R. H.;
Sausville, E. A.; Covell, D. G. Mining the NCI’s tumor screening
database: Identification of compounds with similar cellular activities.
J. Med. Chem. 2002, 45, 818–840.
(22) Web address: http://www.ddl.unimi.it.
(23) Morris, G. M.; Goodsell, D. S.; Huey, R.; Olson, A. J. Distributed
automated docking of flexible ligands to proteins: Parallel applications
of AutoDock 2.4. J. Comp.-Aided Mol. Des. 1996, 10, 293–304.
(24) Solis, F. J.; Wets, R. J. Minimization by random search techniques.
Math. Oper. Res. 1981, 6, 19–30.
JM700964U